When the Going Gets Though: the Impact of Health Shocks on Divorce with Anna Sanz de Galdeano and Daniela Vuri
Does illness break marriages? We analyze how unexpected health shocks— sudden diagnoses of cancer, stroke, or heart attack—affect couple dissolution among older adults (50+). Using longitudinal data from the Health and Retirement Study and a quasi-experimental design comparing affected households to those experiencing similar shocks later, we find that health crises increase divorce probability by 19% relative to the baseline rate. This effect builds gradually rather than emerging immediately. We examine three explanatory channels: mental health deterioration, cognitive decline, and financial strain. The evidence indicates these factors play important roles in understanding how health shocks destabilize marriages in later life.
What Ties Us Together? Explaining Synchronized GDP Volatility with Lorenzo Ductor and Danilo Leiva-León. R&R Economics Letters
Amid heightened policy uncertainty, understanding the drivers of global macroeconomic instability becomes increasingly critical. This paper studies the determinants of output volatility synchronization using data for 42 economies worldwide. We construct a bilateral time-varying index of volatility synchronization and infer its drivers using Bayesian Model Averaging, complemented by Weighted Average Least Squares and LASSO. We find that differences in total factor productivity, interest rate, and fiscal policy volatility robustly explain cross-country synchronization, with important nuances between developed and developing countries. Overall, the results highlight the role of technological divergence and macroeconomic policy uncertainty in shaping the international co-movement of output volatility.
The Hidden Toll of Parental ADRD: Causal Effects on Adult Children
This paper provides the first comprehensive causal analysis of how a parental Alzheimer’s/ADRD diagnosis affects adult children. Using sixteen waves of the Health and Retirement Study (1992–2022) and a stacked difference-in-differences design that exploits the staggered timing of parental diagnoses, I find broad-based disruptions across five outcome domains. The probability of providing personal care rises by 10.4 percentage points, total care hours increase by 159 hours per survey wave, and the depression score worsens by 0.082 standard deviations. Employment falls by 7.0 percentage points and full retirement rises by 8.6 percentage points—consistent with permanent workforce withdrawal. The respondent’s spouse increases care provision by 48.6 hours per wave, documenting a withinhousehold externality. Nursing home placement of the affected parent rises more than 30-fold from a pre-diagnosis rate of 0.6 percent. Short-run effects on household income and wealth are statistically insignificant.